Вы находитесь на странице: 1из 10

Review Annals of Internal Medicine

Narrative Review: Hyperkyphosis in Older Persons

Deborah M. Kado, MD, MS; Katherine Prenovost, PhD; and Carolyn Crandall, MD, MS

Hyperkyphosis is a widely recognized yet largely ignored condition. be independently associated with an increased risk for adverse
Although there are no uniform diagnostic criteria for hyperkyphosis, health outcomes, including impaired pulmonary function, decreased
current studies estimate its prevalence among older adults at 20% physical function capabilities, and future fractures. With the grow-
to 40%. The causes and consequences of hyperkyphosis are not ing older population, we now need research that leads to a deeper
well understood. Some physicians think that fractures cause hy- understanding of the causes, consequences, and treatment of this
perkyphosis and that management strategies should focus solely on common condition.
diagnosis and treatment for osteoporosis. Recent studies, however,
demonstrate that many older adults who are most affected by Ann Intern Med. 2007;147:330-338. www.annals.org
hyperkyphosis do not have vertebral fractures. Hyperkyphosis may For author affiliations, see end of text.

(n 408) or warranting detailed consideration of the orig-

W ith aging, the sagittal convexity of the normal tho-
racic spine, known as kyphosis, tends to progress (1).
Hyperkyphosis, an excess of this process, is often called the
inal article (n 210). We also excluded reports that were
not in English (n 1365) and those that addressed the
dowagers hump. Although not precisely known, the preva- following: 1) primary surgical interventions in children or
lence and incidence of hyperkyphosis in older persons is young adults (n 1386); 2) trauma-induced hyperkypho-
probably between 20% and 40% (2 4). The causes and sis, such as thoracolumbar burst fractures (n 107); 3)
consequences of hyperkyphosis are not well understood. hyperkyphosis caused by chronic diseases, such as Pott dis-
Most clinicians and laypersons assume that hyperkyphosis ease (n 558); and 4) other conditions, such as scoliosis
results from underlying vertebral fractures; however, verte- or kyphoscoliosis (n 1108). As we reviewed the remain-
bral fractures are present in only 36% to 37% of the most ing publications, we paid attention to study design and
severe kyphosis cases (5, 6). Furthermore, hyperkyphosis is sample size, characteristics of the participants, and whether
not simply an undesirable cosmetic consequence of aging. studies addressed factors other than kyphosis that could
It may be associated with several adverse health outcomes, have influenced outcomes.
such as poor pulmonary and physical function.
Our narrative review presents evidence that hyper-
kyphosis cannot be fully explained by osteoporosis and that HOW DOES ONE MEASURE KYPHOSIS?
it is a distinct geriatric syndrome deserving of more atten- The normal spine has 3 curves in the sagittal plane:
tion. With increased clinical awareness informed by con- cervical lordosis (anteriorly convex), thoracic kyphosis (an-
tinued research, physicians can begin to help prevent and teriorly concave), and lumbar lordosis. Kyphosis can be
treat hyperkyphosis and may avoid or lessen attendant ad- measured from radiographs or with such devices as the
verse health consequences. kyphometer (7), goniometer (8), inclinometer (9), and
flexible ruler (1) (Figure 1). Developed first to assess sco-
METHODS liosis angles on spinal radiographs, the Cobb angle was
modified to measure kyphosis. Considered the current gold
We searched MEDLINE and PubMed (OLDMED-
standard measurement, it is calculated by drawing a line at
LINE for pre-1966) for studies from 1950 through 28
the upper border of the vertebral body, marking the begin-
November 2006 by using the following Medical Subject
ning of the thoracic curve (commonly T4), and at the
Heading terms and keywords: kyphosis, hyperkyphosis, and
inferior border of the vertebral body, representing the in-
kyphotic posture. We reviewed all citations and available
terface between the thoracic and lumbar curves (commonly
abstracts (4734 for kyphosis, 77 for hyperkyphosis, and 87 for
T12). Perpendicular lines are drawn from these 2 lines, and
kyphotic posture). We classified 618 abstracts as not relevant
the angle of their intersection is the Cobb angle (Figure 2).
Other clinical assessments of kyphosis include qualitative
visual measurements (3, 4) and measurement of either the
See also: distance from the occiput to wall (10) or the number of
1.7-cm blocks between the head and examination table
while the patient is lying flat with the neck in a neutral
Key Summary Points . . . . . . . . . . . . . . . . . . . . . . . 331
position (2).
Web-Only Research has not fully disentangled the differences be-
CME quiz tween clinical and radiologic kyphosis measures. In a study
Conversion of graphics into slides of 26 postmenopausal women (7), independent observers
Audio summary measured kyphosis similarly when they used Debrunner
kyphometer, flexicurve-derived kyphosis angle, and radio-
330 2007 American College of Physicians
Hyperkyphosis in Older Persons Review

graphic Cobb angle measurements. Intraclass correlation

estimates for the measures ranged from 0.87 to 0.92 (7). Key Summary Points
Traditionally, the Cobb angle is measured from stand- Hyperkyphosis is a common geriatric condition that affects
ing lateral spine films. However, in elderly persons, spinal as many as 20% to 40% of older adults.
radiography is usually performed with the patient in the
supine position for comfort. Because gravitys effect on Hyperkyphosis affects both men and womenthe lay
posture is lost while supine, the kyphosis angle measured in term dowagers hump is a misnomer.
the lying position may underestimate kyphosis. In a study
that compared the standing Debrunner kyphometer angle Vertebral fractures are present in only about one third of
with a supine radiographic Cobb angle measurement in the older adults with the worst degrees of thoracic curva-
120 women age 55 to 80 years (11), the mean difference ture.
between the 2 measures was only 4. The Debrunner
Hyperkyphosis may be associated with multiple adverse
method overestimated the Cobb angle slightly, and the
health outcomes, including impaired pulmonary and physi-
intraclass correlation coefficient for the measures was 0.68
cal function, fractures, and possibly increased risk for

HOW DOES ONE DEFINE HYPERKYPHOSIS? Research about the cause, consequences, and treatment
for hyperkyphosis is urgently needed because available
In younger populations, normal kyphosis angles range
evidence is scant and the prevalence of the condition will
between 20 and 40 (12). In older adults, the mean ky-
probably increase as the population ages.
phosis angle is about 48 to 50 in women (6, 1315) and
about 44 in men (6). We know that the angle increases
with age (1, 6, 12, 16 26), but we do not have uniformly
accepted thresholds for defining either hyperkyphosis or
index increased with age and explained 42% of the varia-
normal thoracic spine changes associated with aging.
tion in kyphosis in men and 48% in women (29). Other
One longitudinal study of 100 healthy men and
studies reported correlations between wedge angles and ky-
women age 50 years or older (mean age, 62 years) reported
phosis ranging from 0.45 to 0.78 (14, 18, 19). In studies
a mean thoracic angle increase of 3 per decade (27). In a
that compared kyphosis between people with and without
longitudinal study of 10 women (mean age, 77 years) fol-
vertebral fractures, those with vertebral fractures had worse
lowed for 3 years after a vertebral fracture (24), the mean
kyphosis (6, 15, 32).
angle increase was 5.6. Cross-sectional studies reported
that the oldest age groups had the most pronounced in- Postural Changes
creases in kyphosis (12, 21, 23). For example, 1 study of Postural changes affecting the cervical, lumbar, and
men and women reported mean thoracic kyphosis angles of sacral spinal areas and postural flexibility may influence
26 in persons in their 20s, 53 in those 60 to 74 years of thoracic curvature. For example, investigators have found
age, and 66 in those older than 75 years of age (23). An weak but statistically significant correlations between Cobb
ongoing large prospective cohort study will soon provide angle thoracic kyphosis and cervical lordosis in men (r
important longitudinal data on rates of progression in 0.27; P 0.03) and women (r 0.33; P 0.009) (23).
older women (28). In 300 volunteers age 20 to 70 years, researchers reported a
correlation of 0.35 between radiographic measures of
thoracic kyphosis and maximum lumbar lordosis (P
POTENTIAL CAUSES OF HYPERKYPHOSIS 0.001) (33). In a study involving 51 women (34), those age
Vertebral fractures do not explain all cases of hyper- 66 to 88 years had greater flexicurve kyphosis than those
kyphosis (29). Only 36% to 37% of older persons with the age 21 to 51 years (mean kyphotic index, 11.1 [SD, 3.9]
worst degrees of kyphosis have underlying vertebral frac- vs. 7.2 [SD, 2.2] during erect stance). The older women
tures (5, 6). Other postulated causes of hyperkyphosis in- were less able to actively correct their posture from a re-
clude postural changes, degenerative disk disease, muscular laxed to an erect position (34).
weakness, ligamentous degeneration, and genetic predispo- Degenerative Disk Disease
sition (Figure 3). A few cross-sectional studies report an association be-
Vertebral Fractures: The Most Cited Cause of tween degenerative disk disease and kyphosis (6, 19, 35).
Hyperkyphosis In a study of 100 healthy women age 39 to 91 years (19),
In 1963, 2 publications noted an association between investigators found a statistically significant correlation be-
vertebral body wedging and kyphosis (30, 31). In the tween anterior disk height and kyphosis angle (r 0.34;
1970s, Milne and Lauder (1) developed the flexicurve mea- P 0.001). Using measurements of lateral spine radio-
surement of kyphosis and quantified the vertebral wedging graphs and mid-sagittal computed tomography films of 93
index from lateral chest radiographs. The wedge deformity ex vivo spines from men and women age 18 to 94 years,
www.annals.org 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 331
Review Hyperkyphosis in Older Persons

Intervertebral Ligaments
Figure 1. The flexicurve: a noninvasive measurement of
Intervertebral ligaments provide stability to the spine
thoracic kyphosis.
and, with aging, are susceptible to loss of elastic tissue,
calcification, and ossification (43). In theory, any of these
changes might predispose an aging individual to hyper-
kyphosis. Indirect evidence that tight ligaments may con-
tribute to hyperkyphosis comes from a study of 11 cadav-
ers, in which segmental transections of the anterior
longitudinal ligament from T3 to T7 resulted in a 16
Cobb angle reduction (44).
Genetic or Metabolic Basis of Hyperkyphosis
Early-onset hyperkyphosis occurs in inherited genetic
conditions, including osteogenesis imperfecta, the Ehlers
Danlos syndrome, the Marfan syndrome, cystic fibrosis,
and Scheuermann disease. Scheuermann disease, the most
frequent cause of juvenile hyperkyphosis, is characterized
by a thoracic Cobb angle of 45 or greater and/or lumbar
kyphosis with associated vertebral wedging, end-plate irreg-
ularity, and disk space narrowing (45). The disease is prob-
ably inherited (46 49). However, with an estimated pop-
Reproduced from Lundon and colleagues (7) with permission of Spine. ulation prevalence of 8.3% (50), it cannot account for the
large prevalence of age-associated hyperkyphosis. Incom-
plete penetrance of certain genotypes may lead to milder
researchers reported a correlation of 0.52 (P 0.001) age-associated hyperkyphosis. In recent studies, older per-
between anterior disk wedging and Cobb angle (35). In a
third study of 1407 older men and women (6), with each
5-increase in Cobb angle, participants had a 36% (95% Figure 2. The Cobb angle of thoracic kyphosis, calculated
CI, 1.26 to 1.48) increased odds of having degenerative from a lateral radiograph.
disk disease. These cross-sectional studies did not establish
whether degenerative disk disease contributed to or was a
consequence of hyperkyphosis.
Muscle Weakness
With 2 exceptions (36, 37), most studies report an
inverse correlation between muscle strength and hyper-
kyphosis (8, 38 42). Although not a direct measure of
spinal muscle strength, grip strength measured in 151 men
and women age 65 to 85 years correlated with worse ky-
phosis measured from a lateral photograph (r 0.25;
P 0.005) (38). Similar findings of upper-extremity
strength and flexicurve-measured kyphosis include a corre-
lation of 0.32 in 47 women volunteers age 50 to 60 years
(P 0.05) (39).
Three studies reported a statistically significant associ-
ation between back-extensor strength and hyperkyphosis in
postmenopausal women. Two of these studies found cor-
relations ranging from 0.30 to 0.35 (40, 41). The third
study found that, among 25 women, those with hyper-
kyphosis (Cobb angle, 50 to 55) had weaker back-exten-
sor muscles (250 N vs. 150 N; P 0.05) than control
patients (Cobb angle 35). No adjustments were made
for apparent differences between the groups in age, body
size, or physical activity (42). In a study of 189 women age
50 to 80 years, those with weak back-extensor muscles
were more likely to have greater kyphosis (58 vs. 51 by
goniometer; P 0.01) (8). Reproduced from Kado and colleagues (11) with permission of Spine.
332 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 www.annals.org
Hyperkyphosis in Older Persons Review

sons who reported a family history of dowagers hump vertebral fractures (P 0.02) but not compared with con-
were statistically significantly more likely to have greater trol patients. No adjustments were made for possible con-
kyphosis, independent of a family history of osteoporosis, founding (67). In 231 community-dwelling volunteers age
bone mineral density, or vertebral fractures (5153). 59 to 89 years, after adjustment for age, sex, body mass
Hyperkyphosis may be an end result of genetic lesions index, moderate to severe scoliosis, heart rate response to
involving genes important for DNA repair and delaying exercise, arthritis, and vertigo, those with hyperkyphosis
senescence. Several mouse knockout models that develop (assessed qualitatively and as the occiput-to-wall distance)
early senescence have pronounced hyperkyphosis (54 59). had reduced gait speed (P 0.015) and increased stair-
Whereas some researchers postulate that hyperkyphosis re- climbing time (P 0.001) (3).
sults from poor posture, results from the mouse studies In a Japanese community sample (n 237), persons
suggest that gravitational forces may not be an important 80 years of age or older with qualitatively assessed thoracic
cause because, unlike humans who stand erect, mice spend hyperkyphosis demonstrated slower gait speed (0.53 m/s
their lives in a prone position. vs. 0.59 m/s; P 0.05), a slower timed get-up-and-go test
result (14.9 seconds vs. 12.5 seconds; P 0.01), and less
functional reach (18.4 cm vs. 23.5 cm; P 0.001) than
ADVERSE HEALTH CONSEQUENCES OF HYPERKYPHOSIS did those classified as having normal posture. No adjust-
Impaired Pulmonary Function ments were made for possible confounding by age and
In a case series of 74 women referred for osteoporosis comorbid conditions (68). Of 1578 older U.S. persons,
evaluation, increased Cobb angle was associated with de- participants with hyperkyphosis defined by blocks had
creased FVC (R2 0.14; P 0.001) (60). Three other worse grip strength and more difficulty rising from a chair
small cross-sectional studies reported statistically significant (odds ratios, 1.5 to 3.7) with increasing kyphosis severity
negative correlations between hyperkyphosis and vital ca- compared with those with normal kyphosis. These findings
pacity and/or FEV1, with values ranging between 0.38 seemed to be independent of age, bone mineral density,
for vital capacity to 0.73 for FEV1 (61 63). In 323 older and self-reported vertebral fractures (69).
community-dwelling persons, those with qualitatively as- Increased Falls
sessed hyperkyphosis had restrictive ventilatory dysfunc-
Some postulate that hyperkyphosis fundamentally al-
tion, even after adjustment for age, sex, and body mass
ters balance and predisposes persons to falls (42, 70, 71).
index (adjusted prevalence odds ratio, 2.3 [CI, 1.1 to
Some studies challenge this assumption. Among 22 women
4.80]; P 0.02). After adjustment for age, sex, smoking,
with osteoporosis diagnosis, there was no association be-
and history of chronic bronchitis or asthma, persons
tween Cobb angle kyphosis and balance measures (72).
with hyperkyphosis were also more likely to have ob-
Similarly, in 151 older men and women, there was no
structive ventilatory patterns (adjusted prevalence odds
association between kyphosis and postural sway (38). Stud-
ratio, 3.3 [CI, 1.7 to 6.5]; P 0.001) and self-reported
ies that have investigated hyperkyphosis and fall risk report
dyspnea (adjusted prevalence odds ratio, 2.5 [CI, 1.1 to
conflicting results (9, 70, 73, 74). In the largest study (n
5.8]) (64).
1883), block-measured hyperkyphosis was statistically sig-
nificantly associated with self-reported injurious falls.
Diminished Physical Function
However, in multivariable analyses, the risk remained sta-
Several cross-sectional studies report an association be-
tistically significant only in men. The study was limited by
tween hyperkyphosis and poor physical function (3, 40,
its retrospective design (74). The association between hyper-
65 69). In an unadjusted analysis of 85 osteoporotic
kyphosis and fall risk warrants further investigation.
women age 50 to 82 years, those with hyperkyphosis re-
ported increased difficulties in bathing and washing (65). Increased Fractures
In another study of 98 postmenopausal women, those with Hyperkyphosis is associated with thoracic vertebral
curviscope-documented hyperkyphosis were more likely to fractures (14, 15, 18, 29). Many presume that fractures
report impaired physical mobility (66). lead to hyperkyphosis. However, with forward-bent pos-
Among studies of performance-based measures, one ture, changes in gravitational loads may increase fracture
involving women age 50 to 60 years found that, with in- risk through associations with gait instability and falls. In a
creasing flexicurve kyphosis, bench-press strength de- large prospective study (75), women with worse Debrun-
creased (r 0.32; P 0.05) and maximum oxygen ner kyphometermeasured kyphosis had more incident
uptake decreased (r 0.37; P 0.02) (39). In a study vertebral fractures. This effect seemed to be largely ac-
of 55 women, those with osteoporosis and vertebral frac- counted for by preexisting prevalent vertebral fracture, a
tures had greater Cobb angle kyphosis (median, 60) than known fracture risk factor (75). In another large prospec-
those with osteoporosis and no vertebral fractures (median, tive study (76), women with self-reported hyperkyphosis
43.5) or selected control patients (median, 37) (67). had a statistically significantly increased risk for future hip
They also had greater oxygen consumption and energy ex- fracture over 3 years (odds ratio, 3.3 [CI, 1.42 to 7.87];
penditure at rest than did those with osteoporosis without P 0.006). Although investigators adjusted for some po-
www.annals.org 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 333
Review Hyperkyphosis in Older Persons

Figure 3. Postulated causes and consequences of hyperkyphosis in older persons.

Multifactorial Health Conditions Adverse Health Conditions

Vertebral fractures Worse pulmonary function

Muscle weakness Poor physical function
Disk height loss Falls Increased
Low bone density Hyperkyphosis Osteoporotic fractures Mortality
Ligament contraction Decreased quality of life
Physical inactivity Depression
Postural changes Functional dependence

tential confounders in a multivariable analysis, they did not Increased Deaths

control for underlying vertebral fractures (76). In a case Some studies suggest associations between hyper-
control study, women with a history of wrist fracture had kyphosis and increased risk for death in older men and
greater thoracic kyphosis than did age-matched control pa- women (2, 5, 29, 92). One study reported that older
tients with no previous wrist, hip, or spine fractures (P women had an age-adjusted, 2.6-fold (CI, 1.3 to 5.1) in-
0.001) (77). Finally, in a study of 596 older women fol- creased risk for a pulmonary-related death with each 1-SD
lowed prospectively over 4 years, women with block- increase in kyphosis, but only 38% of those in the top
defined hyperkyphosis were at a 1.7-fold increased risk for quintile of kyphosis who died a pulmonary-related death
future fracture (CI, 1.0 to 2.97; P 0.049), independent had vertebral fractures (5). Two recent cohort studies con-
of age, bone mineral density, or previous fracture, com- firm that hyperkyphosis, independent of vertebral fractures
pared with those without hyperkyphosis (78). These find- and low bone mineral density, is associated with increased
ings suggest that hyperkyphosis may be associated with risk for deathin particular, with increased risk for pul-
increased fracture risk, independent of osteoporosis. monary death (2, 92).

Other Health-Related Consequences

Age-related hyperkyphosis may be associated with
Standard therapies for age-related hyperkyphosis do
other adverse health effects. Two cross-sectional studies of
not yet exist. A plain spine radiograph can rule out under-
approximately 350 women reported that those with greater
lying osteoporotic vertebral fractures. The presence of frac-
kyphosis had increased odds of having pelvic prolapse (79, tures would justify the use of bisphosphonates or other
80). Case reports have also shown associations between osteoporosis therapy. Approximately 2 of 3 patients with
hyperkyphosis and insufficiency fractures of the sternum hyperkyphosis will not have a vertebral fracture. Those
(81 85) and dysphagia (86). without vertebral fractures should undergo bone density
testing. Low bone mineral density (T-score 2.5) would
justify starting osteoporosis therapy. We do not know
Impact on Quality of Life whether treating low bone mineral density can improve
Of studies that reported associations between hyper- hyperkyphosis, but analyses of data from a large random-
kyphosis and back pain, all but 1 reported statistically sig- ized clinical trial may soon help clarify whether bisphos-
nificant positive associations (15, 65, 8790). Another phonate therapy administration improves kyphosis (93).
study examined osteoporosis and quality of life and re- Nonsurgical Interventions
ported that after adjustment for age, patients with osteo- One randomized trial tested whether spinal orthosis
porosis and self-reported postural changes had increased improves trunk muscle strength, decreases the kyphosis an-
physical difficulty, made more adaptations in their daily gle, and improves quality of life in women with known
life, and had more fears about the future than did those vertebral fractures (94). The investigators reported statisti-
without osteoporosis (91). Demonstrating that the ill ef- cally significant improvement in all measured outcomes
fects of hyperkyphosis are not necessarily due to underlying and an 11% decrease in the kyphosis angle. Major limita-
fractures, women with postural changes and no osteoporo- tions included a small sample size (n 62) and no men-
sis had statistically significantly more physical difficulty tion of whether blinded outcome assessments were made.
and lifestyle adaptations than did those with a previous Another study evaluated the effect of back-strengthening
fracture and no postural changes. exercises on posture in 60 postmenopausal women and
334 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 www.annals.org
Hyperkyphosis in Older Persons Review

reported no overall differences in either back strength or tutes of Healthfunded studies on hyperkyphosis are on-
kyphosis between the intervention and control groups. In a going. One study funded by the National Institutes of
post hoc analysis, women with substantial kyphosis and the HealthNational Institute on Aging prospectively follows
greatest increases in back-extensor strength demonstrated a 1000 women age 65 years or older over 15 years. It aims to
2.8-decrease in kyphosis angle (95). The exercise interven- document changes in thoracic kyphosis curvature, identify
tion was of reasonable duration (2 years), but the study was risk factors for worsening kyphosis, and determine the de-
small, nonblinded measurements were used, and positive gree of kyphosis that may be associated with adverse out-
findings were observed mostly in post hoc subgroup anal- comes (such as functional dependence) (28). Because the
yses. A third trial enrolled 12 women with hyperkyphosis dowagers hump also affects older men, similar studies of
(Cobb angle, 50 to 65; mean age, 77 years) and 13 fe- older men are needed. The second ongoing National Insti-
male control participants (mean age, 71 years). After a tutes of Healthfunded study is a randomized clinical trial
4-week intervention of wearing a spinal-weighted kypho- of a yoga intervention to increase spinal flexibility and de-
orthosis and using a specified back extension and gait pro- crease the kyphosis degree in older adults with hyper-
gram, women with hyperkyphosis demonstrated statisti- kyphosis (102). Results from the ongoing observational
cally significant improvement in balance and gait variables. studies and intervention trials should soon provide impor-
However, the control and intervention groups were of dis- tant information on modifiable risk factors for hyper-
parate ages and kyphosis improvement and control group kyphosis.
outcomes were not reported (96).
Two studies investigated whether spinal flexibility ex- CONCLUSION
ercises reduced hyperkyphosis in older persons (97, 98). A
In the words of Marcel Proust (103): The real voyage
randomized clinical trial of 210 older adults with impaired
of discovery consists not in seeking new landscapes but in
cardiorespiratory fitness that compared spinal flexibility
having new eyes. Hyperkyphosis is an easily recognized
plus aerobic training versus aerobic training only found no
and commonly observed postural change associated with
added benefit of the spinal flexibility training on physical
aging. Yet remarkably, todays physicians have little knowl-
functional outcomes or kyphosis angle (97). The other study
edge about its causes, consequences, or treatment to offer
was of a 3-month yoga intervention in older women with
affected patients. Because the existing overall body of per-
hyperkyphosis (n 21), in which participants demon-
tinent evidence is very limited, we believe that multidisci-
strated improved physical function without kyphosis im-
plinary research involving basic scientists, clinician re-
provement (98). While neither study demonstrated a ky-
searchers, and others is urgently needed to better delineate
phosis benefit, clinically significant hyperkyphosis probably
the causes of age-related hyperkyphosis, confirm and ex-
develops slowly over years. To be successful, an exercise
pand knowledge about its prognostic implications, and de-
intervention probably requires long-term implementation.
velop effective prevention and treatments.
Surgical Interventions
In the setting of painful vertebral fractures, vertebro- From the David Geffen School of Medicine at the University of Califor-
plasty and kyphoplasty are minimally invasive procedures nia, Los Angeles, Los Angeles, California.
that are gaining wider acceptance as effective treatments for
Acknowledgments: The authors thank Paul Mischel, MD, and Arun
intractable pain and possibly kyphosis. Several published
Karlamangla, PhD, MD, for their insightful comments and suggestions,
case series of both procedures and reviews of cases reported and John Ramirez, BA, for his help in preparing the manuscript for
reduced pain, improved physical function, vertebral height publication.
restoration of up to 90%, and kyphosis angle correction
ranging from 8.5 to 14 (99, 100). No published random- Grant Support: By the National Institutes of Health and National In-
ized, controlled trials, however, have evaluated or com- stitute on Aging (grants RO1 AG24246 and 1K12AG01004).
pared benefits and harms of the procedures.
More invasive surgical interventions for hyperkyphosis Potential Financial Conflicts of Interest: Consultancies: D.M. Kado
are associated with high rates of intra- and perioperative (Kyphon).
complications (33%). As such, they are generally reserved
Requests for Single Reprints: Deborah M. Kado, MD, MS, David
for patients with documented curve progression, intracta- Geffen School of Medicine at University of California, Los Angeles,
ble pain, or neurologic compromise (101). 10945 Le Conte Avenue, Suite 2339, Los Angeles, CA 90095; e-mail,
ONGOING RESEARCH Current author addresses are available at www.annals.org.
Most published studies of hyperkyphosis are cross-
sectional and include only a few participants, hampering
causal inference and generalizability. The eRA (Electronic References
Research Administration) Commons Web site (http: 1. Milne JS, Lauder IJ. Age effects in kyphosis and lordosis in adults. Ann Hum
//commons.era.nih.gov) indicates that 2 National Insti- Biol. 1974;1:327-37. [PMID: 4419577]

www.annals.org 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 335
Review Hyperkyphosis in Older Persons

2. Kado DM, Huang MH, Karlamangla AS, Barrett-Connor E, Greendale GA. Puisieux F, et al. Evaluation of spinal curvatures after a recent osteoporotic verte-
Hyperkyphotic posture predicts mortality in older community-dwelling men and bral fracture. Joint Bone Spine. 2002;69:201-8. [PMID: 12027313]
women: a prospective study. J Am Geriatr Soc. 2004;52:1662-7. [PMID: 25. Bartynski WS, Heller MT, Grahovac SZ, Rothfus WE, Kurs-Lasky M.
15450042] Severe thoracic kyphosis in the older patient in the absence of vertebral fracture:
3. Ryan SD, Fried LP. The impact of kyphosis on daily functioning. J Am association of extreme curve with age. AJNR Am J Neuroradiol. 2005;26:2077-
Geriatr Soc. 1997;45:1479-86. [PMID: 9400558] 85. [PMID: 16155162]
4. Takahashi T, Ishida K, Hirose D, Nagano Y, Okumiya K, Nishinaga M, 26. Ostrowska B, Rozek-Mroz K, Giemza C. Body posture in elderly, physically
et al. Trunk deformity is associated with a reduction in outdoor activities of active males. Aging Male. 2003;6:222-9. [PMID: 15006260]
daily living and life satisfaction in community-dwelling older people. Osteoporos 27. Kobayashi T, Atsuta Y, Matsuno T, Takeda N. A longitudinal study of
Int. 2005;16:273-9. [PMID: 15235766] congruent sagittal spinal alignment in an adult cohort. Spine. 2004;29:671-6.
5. Kado DM, Browner WS, Palermo L, Nevitt MC, Genant HK, Cummings [PMID: 15014278]
SR. Vertebral fractures and mortality in older women: a prospective study. Study 28. Kado DM. Causes and consequences of hyperkyphosis in older women: the
of Osteoporotic Fractures Research Group. Arch Intern Med. 1999;159:1215- Study of Osteoporotic Fractures [Grant proposal]. National Institutes of Health/
20. [PMID: 10371229] National Institute on Aging RO1 AG24246. September 2004 June 2008.
6. Schneider DL, von Muhlen D, Barrett-Connor E, Sartoris DJ. Kyphosis does 29. Milne JS, Williamson J. A longitudinal study of kyphosis in older people.
not equal vertebral fractures: the Rancho Bernardo study. J Rheumatol. 2004;31: Age Ageing. 1983;12:225-33. [PMID: 6624608]
747-52. [PMID: 15088302] 30. Nicholas JA, Wilson PD. Osteoporosis of the aged spine. Clin Orthop Relat
7. Lundon KM, Li AM, Bibershtein S. Interrater and intrarater reliability in the Res. 1963;26:19-33. [PMID: 13938329]
measurement of kyphosis in postmenopausal women with osteoporosis. Spine. 31. Rowe CR, Sorbie C. Fractures of the spine in the aged. Clin Orthop Relat
1998;23:1978-85. [PMID: 9779531] Res. 1963;26:34-49. [PMID: 13982805]
8. Mika A, Unnithan VB, Mika P. Differences in thoracic kyphosis and in back 32. Lyles KW, Gold DT, Shipp KM, Pieper CF, Martinez S, Mulhausen PL.
muscle strength in women with bone loss due to osteoporosis. Spine. 2005;30: Association of osteoporotic vertebral compression fractures with impaired func-
241-6. [PMID: 15644764] tional status. Am J Med. 1993;94:595-601. [PMID: 8506884]
9. OBrien K, Culham E, Pickles B. Balance and skeletal alignment in a group of 33. Vialle R, Levassor N, Rillardon L, Templier A, Skalli W, Guigui P. Radio-
elderly female fallers and nonfallers. J Gerontol A Biol Sci Med Sci. 1997;52: graphic analysis of the sagittal alignment and balance of the spine in asymptom-
B221-6. [PMID: 9224428] atic subjects. J Bone Joint Surg Am. 2005;87:260-7. [PMID: 15687145]
10. Leidig-Bruckner G, Minne HW, Schlaich C, Wagner G, Scheidt-Nave C, 34. Hinman MR. Comparison of thoracic kyphosis and postural stiffness in
Bruckner T, et al. Clinical grading of spinal osteoporosis: quality of life compo- younger and older women. Spine J. 2004;4:413-7. [PMID: 15246302]
nents and spinal deformity in women with chronic low back pain and women 35. Goh S, Price RI, Leedman PJ, Singer KP. The relative influence of vertebral
with vertebral osteoporosis. J Bone Miner Res. 1997;12:663-75. [PMID:
body and intervertebral disc shape on thoracic kyphosis. Clin Biomech (Bristol,
Avon). 1999;14:439-48. [PMID: 10521627]
11. Kado DM, Christianson L, Palermo L, Smith-Bindman R, Cummings SR,
36. Miyakoshi N, Hongo M, Maekawa S, Ishikawa Y, Shimada Y, Okada K, et
Greendale GA. Comparing a supine radiologic versus standing clinical measure-
al. Factors related to spinal mobility in patients with postmenopausal osteo-
ment of kyphosis in older women: the Fracture Intervention Trial. Spine. 2006;
porosis. Osteoporos Int. 2005;16:1871-4. [PMID: 15959615]
31:463-7. [PMID: 16481959]
37. Eagan MS, Sedlock DA. Kyphosis in active and sedentary postmenopausal
12. Fon GT, Pitt MJ, Thies AC Jr. Thoracic kyphosis: range in normal subjects.
women. Med Sci Sports Exerc. 2001;33:688-95. [PMID: 11323534]
AJR Am J Roentgenol. 1980;134:979-83. [PMID: 6768276]
38. Brocklehurst JC, Robertson D, James-Groom P. Skeletal deformities in the
13. Cowan NR. The frontal cardiac silhouette in older people. Br Heart J.
elderly and their effect on postural sway. J Am Geriatr Soc. 1982;30:534-8.
1965;27:231-5. [PMID: 14264013]
[PMID: 7096856]
14. De Smet AA, Robinson RG, Johnson BE, Lukert BP. Spinal compression
39. Chow RK, Harrison JE. Relationship of kyphosis to physical fitness and bone
fractures in osteoporotic women: patterns and relationship to hyperkyphosis. Ra-
diology. 1988;166:497-500. [PMID: 3336728] mass on post-menopausal women. Am J Phys Med. 1987;66:219-27. [PMID:
15. Ensrud KE, Black DM, Harris F, Ettinger B, Cummings SR. Correlates of 3434624]
kyphosis in older women. The Fracture Intervention Trial Research Group. J Am 40. Sinaki M, Itoi E, Rogers JW, Bergstralh EJ, Wahner HW. Correlation of
Geriatr Soc. 1997;45:682-7. [PMID: 9180660] back extensor strength with thoracic kyphosis and lumbar lordosis in estrogen-
16. Sidhu LS, Singal P. Age changes in kyphosis and lordosis of Jat-Sikh and deficient women. Am J Phys Med Rehabil. 1996;75:370-4. [PMID: 8873705]
Bania females of Punjab (India) from 20 to 80 years. Anthropol Anz. 1983;41: 41. Sinaki M, Wollan PC, Scott RW, Gelczer RK. Can strong back extensors
59-66. [PMID: 6847160] prevent vertebral fractures in women with osteoporosis? Mayo Clin Proc. 1996;
17. Singer KP, Jones TJ, Breidahl PD. A comparison of radiographic and com- 71:951-6. [PMID: 8820769]
puter-assisted measurements of thoracic and thoracolumbar sagittal curvature. 42. Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR. Balance dis-
Skeletal Radiol. 1990;19:21-6. [PMID: 2326651] order and increased risk of falls in osteoporosis and kyphosis: significance
18. Puche RC, Morosano M, Masoni A, Perez Jimeno N, Bertoluzzo SM, of kyphotic posture and muscle strength. Osteoporos Int. 2005;16:1004-10.
Podadera JC, et al. The natural history of kyphosis in postmenopausal women. [PMID: 15549266]
Bone. 1995;17:239-46. [PMID: 8541136] 43. Resnick D. Degenerative diseases of the vertebral column. Radiology. 1985;
19. Manns RA, Haddaway MJ, McCall IW, Cassar Pullicino V, Davie MW. 156:3-14. [PMID: 3923556]
The relative contribution of disc and vertebral morphometry to the angle of 44. Birnbaum K, Siebert CH, Hinkelmann J, Prescher A, Niethard FU. Cor-
kyphosis in asymptomatic subjects. Clin Radiol. 1996;51:258-62. [PMID: rection of kyphotic deformity before and after transection of the anterior longi-
8617037] tudinal ligamenta cadaver study. Arch Orthop Trauma Surg. 2001;121:142-7.
20. Schenkman M, Shipp KM, Chandler J, Studenski SA, Kuchibhatla M. [PMID: 11262779]
Relationships between mobility of axial structures and physical performance. Phys 45. Ali RM, Green DW, Patel TC. Scheuermanns kyphosis. Curr Opin Pediatr.
Ther. 1996;76:276-85. [PMID: 8602413] 1999;11:70-5. [PMID: 10084088]
21. Korovessis PG, Stamatakis MV, Baikousis AG. Reciprocal angulation of 46. McKenzie L, Sillence D. Familial Scheuermann disease: a genetic and linkage
vertebral bodies in the sagittal plane in an asymptomatic Greek population. study. J Med Genet. 1992;29:41-5. [PMID: 1552543]
Spine. 1998;23:700-4; discussion 704-5. [PMID: 9549792] 47. Axenovich TI, Zaidman AM, Zorkoltseva IV, Kalashnikova EV, Borodin
22. Tuzun C, Yorulmaz I, Cindas A, Vatan S. Low back pain and posture. Clin PM. Segregation analysis of Scheuermann disease in ninety families from Siberia.
Rheumatol. 1999;18:308-12. [PMID: 10468171] Am J Med Genet. 2001;100:275-9. [PMID: 11343318]
23. Boyle JJ, Milne N, Singer KP. Influence of age on cervicothoracic spinal 48. Graat HC, van Rhijn LW, Schrander-Stumpel CT, van Ooij A. Classical
curvature: an ex vivo radiographic survey. Clin Biomech (Bristol, Avon). 2002; Scheuermann disease in male monozygotic twins: further support for the genetic
17:361-7. [PMID: 12084540] etiology hypothesis. Spine. 2002;27:E485-7. [PMID: 12436008]
24. Cortet B, Roches E, Logier R, Houvenagel E, Gaydier-Souquie`res G, 49. Damborg F, Engell V, Andersen M, Kyvik KO, Thomsen K. Prevalence,

336 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 www.annals.org
Hyperkyphosis in Older Persons Review
concordance, and heritability of Scheuermann kyphosis based on a study of twins. women with osteoporosis. J Orthop Sports Phys Ther. 2005;35:452-60. [PMID:
J Bone Joint Surg Am. 2006;88:2133-6. [PMID: 17015588] 16108586]
50. Findlay A, Conner AN, Connor JM. Dominant inheritance of Scheuer- 74. Kado DM, Huang MH, Nguyen CB, Barrett-Connor E, Greendale GA.
manns juvenile kyphosis. J Med Genet. 1989;26:400-3. [PMID: 2738903] Hyperkyphotic posture and risk of injurious falls in older persons: the rancho
51. Kado DM, Palermo L. Is there evidence of heritability of hyperkyphosis in bernardo study. J Gerontol A Biol Sci Med Sci. 2007;62:652-7. [PMID:
older women? [Abstract]. J Bone Miner Res. 2005;20:S231. 17595423]
52. Huang MH, Barrett-Connor E, Kado DM. Heritability of hyperkyphosis in 75. Shipp KM, Guess HA, Ensrud KE, Nevitt MC, Kado DM, Cummings SR.
older persons [Abstract]. J Gen Intern Med. 2006;21:63-4. Thoracic kyphosis and rate of incident vertebral fracture [Abstract]. J Bone Miner
53. Kado DM, Huang HM, Barrett-Connor E. A maternal history of stooped Res. 2002;17:S174.
posture is a risk factor for hyperkyphosis [Abstract]. J Bone Miner Res. 2006;21: 76. McGrother CW, Donaldson MM, Clayton D, Abrams KR, Clarke M.
S239. Evaluation of a hip fracture risk score for assessing elderly women: the Melton
54. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H, Suga T, Utsugi T, Osteoporotic Fracture (MOF) study. Osteoporos Int. 2002;13:89-96. [PMID:
et al. Mutation of the mouse klotho gene leads to a syndrome resembling ageing. 11883411]
Nature. 1997;390:45-51. [PMID: 9363890] 77. Porter RW, Johnson K, McCutchan JD. Wrist fracture, heel bone density
55. Vogel H, Lim DS, Karsenty G, Finegold M, Hasty P. Deletion of Ku86 and thoracic kyphosis: a case control study. Bone. 1990;11:211-4. [PMID:
causes early onset of senescence in mice. Proc Natl Acad Sci U S A. 1999;96: 2202357]
10770-5. [PMID: 10485901] 78. Huang MH, Barrett-Connor E, Greendale GA, Kado DM. Hyperkyphotic
56. Tyner SD, Venkatachalam S, Choi J, Jones S, Ghebranious N, Igelmann posture and risk of future osteoporotic fractures: the Rancho Bernardo study. J
H, et al. p53 mutant mice that display early ageing-associated phenotypes. Na- Bone Miner Res. 2006;21:419-23. [PMID: 16491290]
ture. 2002;415:45-53. [PMID: 11780111] 79. Lind LR, Lucente V, Kohn N. Thoracic kyphosis and the prevalence of
57. de Boer J, Andressoo JO, de Wit J, Huijmans J, Beems RB, van Steeg H, advanced uterine prolapse. Obstet Gynecol. 1996;87:605-9. [PMID: 8602316]
et al. Premature aging in mice deficient in DNA repair and transcription. Science. 80. Mattox TF, Lucente V, McIntyre P, Miklos JR, Tomezsko J. Abnormal
2002;296:1276-9. [PMID: 11950998] spinal curvature and its relationship to pelvic organ prolapse. Am J Obstet Gy-
58. Razzaque MS, Sitara D, Taguchi T, St-Arnaud R, Lanske B. Premature necol. 2000;183:1381-4; discussion 1384. [PMID: 11120500]
aging-like phenotype in fibroblast growth factor 23 null mice is a vitamin D- 81. Cooper KL. Insufficiency fractures of the sternum: a consequence of thoracic
mediated process. FASEB J. 2006;20:720-2. [PMID: 16436465] kyphosis? Radiology. 1988;167:471-2. [PMID: 3357957]
59. Kakoki M, Kizer CM, Yi X, Takahashi N, Kim HS, Bagnell CR, et al. 82. Chen C, Chandnani V, Kang HS, Schils JP, Resnick D, Bjorkengren AG,
Senescence-associated phenotypes in Akita diabetic mice are enhanced by absence et al. Insufficiency fracture of the sternum caused by osteopenia: plain film find-
of bradykinin B2 receptors. J Clin Invest. 2006;116:1302-9. [PMID: 16604193] ings in seven patients. AJR Am J Roentgenol. 1990;154:1025-7. [PMID:
60. Leech JA, Dulberg C, Kellie S, Pattee L, Gay J. Relationship of lung func- 2108537]
tion to severity of osteoporosis in women. Am Rev Respir Dis. 1990;141:68-71. 83. Sapherson DA, Mitchell SC. Atraumatic sternal fractures secondary to os-
[PMID: 2297189] teoporosis. Clin Radiol. 1990;42:250-1. [PMID: 2225729]
61. Schlaich C, Minne HW, Bruckner T, Wagner G, Gebest HJ, Grunze M, 84. Berg EE. The sternal-rib complex. A possible fourth column in thoracic spine
et al. Reduced pulmonary function in patients with spinal osteoporotic fractures. fractures. Spine. 1993;18:1916-9. [PMID: 8235883]
Osteoporos Int. 1998;8:261-7. [PMID: 9797911] 85. Klaase JM, Zimmerman KW, Veldhuis EF. Increased kyphosis by a combi-
62. Culham EG, Jimenez HA, King CE. Thoracic kyphosis, rib mobility, and nation of fractures of the sternum and thoracic spine. Eur Spine J. 1998;7:69-71.
lung volumes in normal women and women with osteoporosis. Spine. 1994;19: [PMID: 9548363]
1250-5. [PMID: 8073317] 86. Goyal N, Narlawar R, Garrett A. Kyphosis, a rare cause of dysphagia. Age
63. Lombardi I Jr, Oliveira LM, Mayer AF, Jardim JR, Natour J. Evaluation of Ageing. 2005;34:521-2. [PMID: 16107461]
pulmonary function and quality of life in women with osteoporosis. Osteoporos 87. Finsen V. Osteoporosis and back pain among the elderly. Acta Med Scand.
Int. 2005;16:1247-53. [PMID: 15806323] 1988;223:443-9. [PMID: 2967621]
64. Di Bari M, Chiarlone M, Matteuzzi D, Zacchei S, Pozzi C, Bellia V, et al. 88. Ettinger B, Black DM, Palermo L, Nevitt MC, Melnikoff S, Cummings
Thoracic kyphosis and ventilatory dysfunction in unselected older persons: an SR. Kyphosis in older women and its relation to back pain, disability and os-
epidemiological study in Dicomano, Italy. J Am Geriatr Soc. 2004;52:909-15. teopenia: the study of osteoporotic fractures. Osteoporos Int. 1994;4:55-60.
[PMID: 15161454] [PMID: 8148573]
65. Ryan PJ, Blake G, Herd R, Fogelman I. A clinical profile of back pain and 89. Korovessis P, Stamatakis M, Baikousis A. Segmental roentgenographic anal-
disability in patients with spinal osteoporosis. Bone. 1994;15:27-30. [PMID: ysis of vertebral inclination on sagittal plane in asymptomatic versus chronic low
8024847] back pain patients. J Spinal Disord. 1999;12:131-7. [PMID: 10229527]
66. Cortet B, Houvenagel E, Puisieux F, Roches E, Garnier P, Delcambre B. 90. Miyakoshi N, Itoi E, Kobayashi M, Kodama H. Impact of postural defor-
Spinal curvatures and quality of life in women with vertebral fractures secondary mities and spinal mobility on quality of life in postmenopausal osteoporosis.
to osteoporosis. Spine. 1999;24:1921-5. [PMID: 10515017] Osteoporos Int. 2003;14:1007-12. [PMID: 14557854]
67. Lombardi I Jr, Oliveira LM, Monteiro CR, Confessor YQ, Barros TL, 91. Martin AR, Sornay-Rendu E, Chandler JM, Duboeuf F, Girman CJ, Del-
Natour J. Evaluation of physical capacity and quality of life in osteoporotic mas PD. The impact of osteoporosis on quality-of-life: the OFELY cohort. Bone.
women. Osteoporos Int. 2004;15:80-5. [PMID: 14593454] 2002;31:32-6. [PMID: 12110409]
68. Hirose D, Ishida K, Nagano Y, Takahashi T, Yamamoto H. Posture of the 92. Kado DM, Greendale GA, Lui L, Ensrud KE, Fink HA, Hillier T, et al.
trunk in the sagittal plane is associated with gait in community-dwelling elderly Hyperkyphosis predicts mortality independent of vertebral osteoporosis in older
population. Clin Biomech (Bristol, Avon). 2004;19:57-63. [PMID: 14659931] women [Abstract]. J Bone Miner Res. 2003;18:S249.
69. Kado DM, Huang MH, Barrett-Connor E, Greendale GA. Hyperkyphotic 93. Black DM, Reiss TF, Nevitt MC, Cauley J, Karpf D, Cummings SR.
posture and poor physical functional ability in older community-dwelling men Design of the Fracture Intervention Trial. Osteoporos Int. 1993;3 Suppl 3:S29-
and women: the Rancho Bernardo study. J Gerontol A Biol Sci Med Sci. 2005; 39. [PMID: 8298200]
60:633-7. [PMID: 15972617] 94. Pfeifer M, Begerow B, Minne HW. Effects of a new spinal orthosis on
70. Woodhull-McNeal AP. Changes in posture and balance with age. Aging posture, trunk strength, and quality of life in women with postmenopausal
(Milano). 1992;4:219-25. [PMID: 1420405] osteoporosis: a randomized trial. Am J Phys Med Rehabil. 2004;83:177-86.
71. Lynn SG, Sinaki M, Westerlind KC. Balance characteristics of persons with [PMID: 15043351]
osteoporosis. Arch Phys Med Rehabil. 1997;78:273-7. [PMID: 9084349] 95. Itoi E, Sinaki M. Effect of back-strengthening exercise on posture in healthy
72. Greig AM, Bennell KL, Briggs AM, Wark JD, Hodges PW. Balance im- women 49 to 65 years of age. Mayo Clin Proc. 1994;69:1054-9. [PMID:
pairment is related to vertebral fracture rather than thoracic kyphosis in individ- 7967758]
uals with osteoporosis. Osteoporos Int. 2007;18:543-51. [PMID: 17106784] 96. Sinaki M, Brey RH, Hughes CA, Larson DR, Kaufman KR. Significant
73. Arnold CM, Busch AJ, Schachter CL, Harrison L, Olszynski W. The reduction in risk of falls and back pain in osteoporotic-kyphotic women through
relationship of intrinsic fall risk factors to a recent history of falling in older a Spinal Proprioceptive Extension Exercise Dynamic (SPEED) program. Mayo

www.annals.org 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 337
Review Hyperkyphosis in Older Persons

Clin Proc. 2005;80:849-55. [PMID: 16007888] 100. Heini PF. The current treatmenta survey of osteoporotic fracture treat-
97. Morey MC, Schenkman M, Studenski SA, Chandler JM, Crowley GM, ment. Osteoporotic spine fractures: the spine surgeons perspective. Osteoporos
Sullivan RJ Jr, et al. Spinal-flexibility-plus-aerobic versus aerobic-only training: Int. 2005;16 Suppl 2:S85-92. [PMID: 15365699]
effect of a randomized clinical trial on function in at-risk older adults. J Gerontol 101. Heary RF. Evaluation and treatment of adult spinal deformity. Invited
A Biol Sci Med Sci. 1999;54:M335-42. [PMID: 10462164] submission from the Joint Section Meeting on Disorders of the Spine and Pe-
98. Greendale GA, McDivit A, Carpenter A, Seeger L, Huang MH. Yoga for ripheral Nerves, March 2004. J Neurosurg Spine. 2004;1:9-18. [PMID:
women with hyperkyphosis: results of a pilot study. Am J Public Health. 2002; 15291014]
92:1611-4. [PMID: 12356608] 102. Greendale GA. The yoga for hyperkyphosis trial [Grant proposal]. National
99. Hulme PA, Krebs J, Ferguson SJ, Berlemann U. Vertebroplasty and kypho- Institutes of Health/National Institute of Child Health and Human Develop-
plasty: a systematic review of 69 clinical studies. Spine. 2006;31:1983-2001. ment R01 HD045834. 9 September 2004 30 June 2008.
[PMID: 16924218] 103. Proust M. Remembrances of Things Past. Volume 7. 1925.

Free CME
to Subscribers
and Members


338 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 www.annals.org
Annals of Internal Medicine
Current Author Addresses: Drs. Kado and Prenovost: Division of Ge-
riatrics, Department of Medicine, David Geffen School of Medicine at
University of California, Los Angeles, 10945 Le Conte Avenue, Suite
2339, Los Angeles, CA 90095.
Dr. Crandall: Division of General Internal Medicine, Department of
Medicine, David Geffen School of Medicine at University of California,
Los Angeles, 911 Broxton Avenue, 1st Floor, Los Angeles, CA 90024.

www.annals.org 4 September 2007 Annals of Internal Medicine Volume 147 Number 5 W-63